So, I've read the article, and it seems to not support this assertion of yours:
> TLDR: We're undercounting deaths due to covid-19 by ~60% nationally, and 40% in NYC.
What it does say is this:
> “Thus I would probably add about 15% or so to the known death toll,” she wrote to two city officials. “However — no city or state will be factoring this in or reporting it, so I don’t think we should either. We should just assume that the deaths are about 15% more than we can count, but not include them in official modeling, because we will never really know.”
It does pay some lip service to the notion that some of these at-home deaths could be precisely because of the "elective" (meaning, scheduled) surgeries that we had postponed.
Interestingly per the https://www.cdc.gov/heartdisease/facts.htm, we see something like 647,000 heart disease deaths per year. So, I haven't done the napkin math, but could the 20%-odd spike of cardiac arrest events they mentioned have more to do with postponed surgeries?
The link you posted shows 11k/16k deaths from COVID. Nothing about 27k. And the chart I posted doesn't show total deaths in NYC, but some selection of morbidities. Is it wrong? Probably. But official data also says heart attacks and strokes are massively down (60-80%), and covid is way up, and there are big incentive problems. I stand by the point.
>> “Thus I would probably add about 15% or so to the known death toll,” she wrote to two city officials.
There's a few different numbers and it's important to keep them clear.
The main number that people know is "died after being tested positive with covid-19 from a covid-19 like illness". For a while this number only included people who died in hospital.
The next number (which the 40% to 60% refers to) is "died from a covid-19 related cause after testing positive, or after being suspected of being covid-19 positive". This includes all the frail people in nursing homes where tests were not carried out (they weren't being tested because it wasn't going to affect their treatment).
The 15% appears to refer to all the people who live alone at home.
> Interestingly per the https://www.cdc.gov/heartdisease/facts.htm, we see something like 647,000 heart disease deaths per year. So, I haven't done the napkin math, but could the 20%-odd spike of cardiac arrest events they mentioned have more to do with postponed surgeries?
I agree that cancelled surgeries, and also people delaying seeking help when they start to have heart problems, is really worrying and has caused some of this excess death.
>but it's also a methodology that should be likely to give a high side estimate
It's the same methodology they use for all diseases. If it's too high an estimate, they'd pretty quickly find it since it would overcount enough reasons fro death that totals would not match measured death totals.
In my original post I linked evidence and the expert reasons why counting is likely low.
>Outside of New York/New Jersey and Massachusetts, I don't find a high disparity in the excess death ratio that supports the idea that there is significant under reporting of deaths.
The CDC link had many states with many thousands of excess deaths. And remember CDC death count lags actual death counts by a few weeks, so the 118k-154k (95%-ile) they list is to be compared to COVID as of a few weeks ago, which was under 110K two weeks ago. So taking the median gives around 136K actual deaths (prob not all COVID, but majority so) vs 110K reported - approx 20% undercounted.
>So are more people having heart attacks? Is Covid causing heart attacks?
Both of those are likely true [1]. Research since then is looking like COVID is killing asymptomatic people via attacking the heart.
>I don't feel comfortable pushing the idea that we are under-reporting deaths or over-reporting deaths because I think that we don't actually know
I've found significant decent groups and projects claiming undercounting, many with decent empirical evidence. I've not found the opposite, despite looking. I posted a few to get you started - discount as you wish.
I don't know, there's no paper, it's just powerpoint slides.
I have previously poked around NYC-area death statistics and it looks like 1) a massive spike in deaths by any cause that looks like several 9/11s followed by 2) a regression to the usual rate of deaths once COVID was largely brought to heel in the summer months. I can't imagine that the total number of deaths in NYC this year could possibly be the same as in 2019.
But I'm not an expert.
Ed: glancing over the CDC statistics, when COVID-attributed deaths go up, all-cause mortality goes up, too... by approximately the same amount. If a large proportion of those COVID deaths were mislabeled heart disease deaths, why is that? I made these graphs with the same CDC data:
It's amazing how often death toll related to COVID-19 is taken in a vacuum. From all the little evidence we get (https://www.nytimes.com/2020/04/06/well/live/coronavirus-doc...), it seems that heart disease related deaths are reduced, or rather "swallowed" by the coronavirus deaths (in US alone, 650k people die of heart-related diseases every year, or more than 50k a month).
To me, the explanation is logical — some who would have died of a heart disease anyway are now dying of COVID19.
I do not want to dispute that it is having an effect — I just want to know what that effect is. Is it 20% more deaths? Or 4x (300%) more?
Considering heart related diseases have been mostly linked to COVID19 deaths, I am surprised nobody is looking at aggregated data yet, which would help more "accurately" estimate the death toll.
The Washington Post had a good article a while ago on the issues with estimating death rates. The math is a bit complex, so I'll let them explain it to you (instead of a HN comment box). It's nuanced, as these pandemic things tend to be.
TLDR: We're undercounting deaths due to covid-19 by ~60% nationally, and 40% in NYC.
OP here, good point. I posted this as it agreed with the Bergamo data where all-cause mortality was 3x the base line number and only 1x of that was attributed to Covid-19. NYC is shaping up to be similar unfortunately.
So agree, numbers don’t lie. But... handling emergencies effectively can lead to problematic numbers.
My brother and I discussed this at length.
He claims (and I have no reason to doubt) that New York has 5x the usual death rate right now. About 2x the usual death rate is clearly COVID-19, another 1x is probably COVID-19 but not tested, another 1x is unclear and the rest are people who likely would have died anyways.
An example of that unclear bucket are people dying from heart attacks. COVID-19 stresses the heart which can cause that. However thanks to COVID-19, EMTs have moved to "Do not do CPR and if we cannot revive, do not take to the hospital." This procedure makes sense because CPR on a COVID-19 patient sprays the whole room with COVID-19. And taking someone with it to a hospital likely will cause them to get it, and it is particularly bad in people with heart problems.
However it also means that heart attacks have a higher likelihood of killing you than usually.
So..people die of heart attacks. Is that heart attack attributable to COVID-19? Is it fatal just because COVID-19 has overwhelmed the health system? We don't know and aren't trying to find out because knowing isn't the priority in a world where our health care system is stretched to the breaking point. All we really know is that people are dying.
"If you change methodology, you really have to go back and recast historic numbers. But that isn't being done."
That is completely inaccurate. The deaths were added to historic numbers, and the "garbage number" is what was added to the total cumulatively to-date, and did not represent a 24 hour number on official stats. See the graphs on these pages. As you can see there is no discontinuous jump.
Not all the deaths are tested for COVID. Those that aren't don't make it into the stats. [1]
Overall mortality, compared to the baseline, has skyrocketed in the past few weeks. For the March 4 - April 4 time period, it was double the usual rate. [2]
Cool! I like numbers, too. How about this: NYC = 8.8 million people [0], U.S. death rate is currently 8.997 per 1000 [1], which works out to 216 people dying per day in NYC. The current 4 week average for covid deaths in NYC is 5 per day [2]. Hospitalizations are increasing, the 28 day average of 70 patients has increased to 88 in the last 7 days.
5 out of 200+ deaths per day and around 100 beds taken by covid patients doesn't sound like a crisis or justification for this executive order. Maybe I'm missing something?
I also found it interesting that the per-capita deaths this year and last don't show an increase over the prior/predicted line on the graph. I assume this is because the baby boomer generation is getting older.
None of this runs counter to the numbers you presented, but does seem to show that this pandemic is not killing people or putting them in the hospital at a rate anywhere close to what it was doing last year. It's predicted that this will continue to affect people less as it becomes endemic, like the other coronaviruses that circulate and don't bother anyone much [3].
Just because they are counting probable deaths does not mean they aren't also undercounting the true deaths. Given the rate of testing confirmed is clearly << true, so I don't see why there is so much objection to using some form of probable deaths. And I haven't seen any legit claims that the specific mechanism for determining probable deaths used in NYC is net overcounting.
Excess death data so far suggests that they are indeed undercounting true deaths. While it is certainly possible lockdown has caused some additional immediate deaths, it is also likely other deaths (eg decreased car accidents) have gone down. So it will require further analysis once this is all said and done and more data is available, but my money is that the US has undercounted deaths thus far.
> Ascribing all excess deaths to coronavirus is a mistake.
I worded my statement intentionally.
> It's also most likely still an undercount
Some of the excess deaths will be other things. It will take more careful analysis to come up with reasonable breakdowns, and what percentage is likely covid-19 deaths.
> “I think people need to be aware that the data they’re seeing on deaths is very incomplete,” said Dan Weinberger, a Yale professor of epidemiology who led the analysis for The Post.
> Those excess deaths — the number beyond what would normally be expected for that time of year — are not necessarily attributable directly to covid-19, the disease caused by the coronavirus. They could include people with unrelated maladies who avoided hospitals for fear of being exposed or who couldn’t get the care they needed from overwhelmed health systems, as well as some number of deaths that are part of the ordinary variation in the death rate. The number is affected by increases or decreases in other categories of deaths, such as traffic fatalities and homicides.
> But excess deaths are a starting point for scientists to assess the overall impact of the pandemic.
> Serious question, how many of those people would have died in the past 12 months if they didn't get COVID? The number is most likely not 500,000, but it also probably isn't 0.
This can be answered using “excess mortality” data.
Studies have shown during that time period, about 600,000 more people died than would have been expected compared to deaths in the previous year.
The official covid death count is likely an undercount of deaths caused by covid, because a number of people died from covid related complications and got recorded as pneumonia, Alzheimer’s, or heart disease/attack/stroke related deaths.
1. My link stated that in certain areas, such as Connecticut, traffic fatalities went up.
2. My overall point is that a certain number of excess deaths are atrributable to the lockdown itself, rather than Covid. IIRC, the total number of reported deaths in NYC from heart disease, heart attacks, cancer, and a whole bunch of other diseases went down drastically; either Covid-19 magically cures all these other ailments, or a bunch of deaths are being mislabelled as Covid-related.
> TLDR: We're undercounting deaths due to covid-19 by ~60% nationally, and 40% in NYC.
What it does say is this:
> “Thus I would probably add about 15% or so to the known death toll,” she wrote to two city officials. “However — no city or state will be factoring this in or reporting it, so I don’t think we should either. We should just assume that the deaths are about 15% more than we can count, but not include them in official modeling, because we will never really know.”
It does pay some lip service to the notion that some of these at-home deaths could be precisely because of the "elective" (meaning, scheduled) surgeries that we had postponed.
Interestingly per the https://www.cdc.gov/heartdisease/facts.htm, we see something like 647,000 heart disease deaths per year. So, I haven't done the napkin math, but could the 20%-odd spike of cardiac arrest events they mentioned have more to do with postponed surgeries?
reply